diff --git a/deploy/iso/minikube-iso/package/docker-bin/docker-bin.hash b/deploy/iso/minikube-iso/package/docker-bin/docker-bin.hash
index 12c6284338ac..dd19a3287665 100644
--- a/deploy/iso/minikube-iso/package/docker-bin/docker-bin.hash
+++ b/deploy/iso/minikube-iso/package/docker-bin/docker-bin.hash
@@ -1 +1,2 @@
 sha256 893e3c6e89c0cd2c5f1e51ea41bc2dd97f5e791fcfa3cee28445df277836339d  docker-1.11.1.tgz
+md5 1058a7eb3f05833475fcc29fc4b11011  docker-17.05.0-ce.tgz
diff --git a/deploy/iso/minikube-iso/package/docker-bin/docker-bin.mk b/deploy/iso/minikube-iso/package/docker-bin/docker-bin.mk
index 731124c289c7..18fbbba70de1 100644
--- a/deploy/iso/minikube-iso/package/docker-bin/docker-bin.mk
+++ b/deploy/iso/minikube-iso/package/docker-bin/docker-bin.mk
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-DOCKER_BIN_VERSION = 1.11.1
+DOCKER_BIN_VERSION = 17.05.0-ce
 DOCKER_BIN_SITE = https://get.docker.com/builds/Linux/x86_64
 DOCKER_BIN_SOURCE = docker-$(DOCKER_BIN_VERSION).tgz
 
@@ -32,6 +32,10 @@ define DOCKER_BIN_INSTALL_TARGET_CMDS
 	$(INSTALL) -D -m 0755 \
 		$(@D)/docker-containerd-ctr \
 		$(TARGET_DIR)/bin/docker-containerd-ctr
+
+	$(INSTALL) -D -m 0755 \
+		$(@D)/dockerd \
+		$(TARGET_DIR)/bin/dockerd
 endef
 
 define DOCKER_BIN_INSTALL_INIT_SYSTEMD
diff --git a/deploy/iso/minikube-iso/package/docker-bin/docker.service b/deploy/iso/minikube-iso/package/docker-bin/docker.service
index dd6e2ba283f1..fd685676ba85 100644
--- a/deploy/iso/minikube-iso/package/docker-bin/docker.service
+++ b/deploy/iso/minikube-iso/package/docker-bin/docker.service
@@ -13,7 +13,7 @@ Environment=DOCKER_RAMDISK=yes
 # the default is not to use systemd for cgroups because the delegate issues still
 # exists and systemd currently does not support the cgroup feature set required
 # for containers run by docker
-ExecStart=/usr/bin/docker daemon -H fd://
+ExecStart=/usr/bin/dockerd -H fd://
 ExecReload=/bin/kill -s HUP $MAINPID
 # Having non-zero Limit*s causes performance problems due to accounting overhead
 # in the kernel. We recommend using cgroups to do container-local accounting.